Species Names in the PhyloCode: The Approach Adopted by the International Society for Phylogenetic Nomenclature

On May 24, 2007, the Committee on Phylogenetic Nomenclature (CPN), which consists of 12 elected members from the International Society for Phylogenetic Nomenclature (ISPN), adopted a new article in the International Code of Phylogenetic Nomenclature (ICPN or PhyloCode; Cantino and de Queiroz, 2007) addressing the naming of species in the context of phylogenetic nomenclature. This vote, which took place after more than 10 years of discussion on how to handle species names in phylogenetic nomenclature, represents a major step in the development of the PhyloCode. Until now, the successive drafts of the PhyloCode have only dealt with clade names, although the application of phylogenetic nomenclature to species has been heavily debated at workshops and symposia on phylogenetic nomenclature (e.g., Cantino et al., 1999a), in the literature (e.g., de Queiroz and Gauthier, 1992; Graybeal, 1995; Schander and Thollesson, 1995; Cantino, 1998; Cantino et al., 1999a, 1999b; Ereshefsky, 1999; Pleijel, 1999; Pleijel and Rouse, 2000a, 2000b, 2003; Artois, 2001; Hillis et al., 2001; Lee, 2002; Spangler, 2003; Dayrat et al., 2004; Dayrat, 2005; Dayrat and Gosliner, 2005; Fisher, 2006; Wolsan, 2007a, 2007b), and at the two meetings of the ISPN (Laurin and Cantino, 2004, 2006; see the Preface to the PhyloCode for additional information). The article on species names (Article 21) that the CPN recently adopted was prepared by the four of us. Here, we wish to explain its rationale and advantages. The Linnaean binominal nomenclature used for species names in the rank-based codes (Bacterial Code [BC], International Code of Botanical Nomenclature [ICBN], International Code of Zoological Nomenclature [ICZN]) is not fully compatible with phylogenetic nomenclature because their requirements regarding ranks differ. Under the rank-based codes, the name of a species is a combination of two words, i.e., a binomen (or binomial), the first part of which is a generic name and the second part of which is a specific name (ICZN) or epithet (BC, ICBN). The first part of the binomen requires reference to the genus rank, and the second part requires reference to the species rank. The rank-based codes currently allow no exceptions. Contrary to a common misconception, ranks are not prohibited in the PhyloCode, but a taxon name is not based on its rank (if any), and use of a particular rank cannot be mandatory. Thus, it is not the use of the genus rank per se that is incompatible with the PhyloCode, nor is it the use of binomina, but the mandatory use of the genus rank in order to name species and the fact that the species binomen is influenced by ranking (i.e., the first part of the species binomen depends on which taxon is ranked as a genus). Although this incompatibility prevents the combining of Linnaean binomina and phylogenetic nomenclature seamlessly into a single integrated system, Linnaean binomina can be reinterpreted and used in the context of phylogenetic nomenclature in the manner described below. Beyond phylogenetic nomenclature, the mandatory use of genus names sometimes conflicts with the Hennigian principle that all taxa (or at least supraspecific taxa) must be monophyletic. Systematists currently have to assign every species to a particular genus, even in cases where species cannot be referred with confidence to any clade that has been ranked as a genus (Cantino et al., 1999b; Dayrat 2005; Dayrat and Gosliner, 2005). When naming new species in such situations of phylogenetic uncertainty, there are only two solutions under the rank-based codes, neither of which is desirable from a phylogenetic standpoint: (1) to erect new monotypic genera or (2) to assign species to existing nonmonophyletic or questionably monophyletic genera. Under the former approach, a new nomenclatural entity is created that provides no information about the relationships of the species. Under the latter approach, use of generic names that are known to apply to nonmonophyletic or questionably monophyletic taxa does not accurately communicate phylogenetic information and may perpetuate the misconception that these taxa are monophyletic. In consideration of these problems, two divergent approaches to species names have been advocated by proponents of phylogenetic nomenclature who also want to use species names: either develop a mechanism to extend phylogenetic nomenclature to species (e.g., Cantino et al., 1999a; Clarke, 2004; Dayrat, et al., 2004) or restrict the PhyloCode to clade names, leaving the governance of species names to the rank-based codes (Lee, 2002). Some proponents of phylogenetic nomenclature reject species altogether and would therefore restrict biological nomenclature to clades (Mishler, 1999; Pleijel, 1999; Pleijel and Rouse, 2000a, 2000b; Fisher, 2006), a viewpoint that we do not share. The approach that we ultimately adopted is closest to Lee's (2002) but with some important differences that evolved in the process of drafting an entire code for species names—a document that we, its authors, eventually rejected. A set of rules governing species names was included in the first draft of the PhyloCode, prepared by Philip Cantino and Kevin de Queiroz for the first Workshop on Phylogenetic Nomenclature (Harvard University, August 1998). However, the 27 participants attending the workshop could not reach a consensus on the form that species names should take. Considerable discussion and a vote during the workshop revealed a deep division among the participants on whether species names under the PhyloCode should be binomen based or epithet based. In binomen-based methods, species names take the form of a unique binomen, though the first part of the name has no necessary association with the rank of genus; the methods vary in whether the first part of the name can change with changing ideas about relationship (as opposed to being permanently fixed) and in whether the two parts are separated by a symbol (e.g., a hyphen). In epithet-based methods, species names take the form of a uninomen, which, in the case of conversion from existing Linnaean binomina, consists of the specific name (ICZN) or epithet (BC, ICBN) without the genus name. Because uninominal names are not necessarily unique, the uninomen in most epithet-based methods is combined with a number (as part of the name) and/or associated (not as part of the name) with either a registration number or the names of one or more including clades. The distinctions among these various options have been discussed in detail by Cantino et al. (1999a). The deep disagreement on the form of species names at the Harvard workshop led to a decision to restrict the PhyloCode temporarily to clade names, and the first publicly available version of the PhyloCode (April 2000; available at http://www.phylocode.org) dealt only with clade names. The governance of species names was again a major topic of discussion at the Second Workshop on Phylogenetic Nomenclature (Yale University, July 2002), where opinions ranged widely: some participants arguing that species names should never be governed by the PhyloCode, and others maintaining that their inclusion is so critical that the code should not be implemented until expanded to cover species. The majority held the intermediate position that species names should eventually be included in the PhyloCode, but that the controversy surrounding species names should not delay implementation of the rules for clade names. Thus, it was decided at the 2002 workshop that rules for clade names and for species names would be published in separate documents and that the timing of implementation of the two documents would be independent, with implementation of the rules for clade names likely preceding those for species names. Apart from these practical questions, the period from 1998 to 2003 saw intense theoretical and philosophical discussions on the status of species in phylogenetic nomenclature. Mishler (1999) argued that species are not fundamentally distinct from clades, that the species is simply a rank, and that species should not therefore be treated any differently than other clades nomenclaturally. Along the same lines, Pleijel (1999) and Pleijel and Rouse (2000a, 2000b, 2003) argued against the comparability among, and utility species taxa and the for the least clades to which can be In de Queiroz argued that species and clades are fundamentally of the former being a of a single a clade all of the of from a species and clades are of also should be the of the ISPN July for a species code was as a of ideas by Dayrat and The majority of the participants a for development of a species separate from but compatible with the code for clade names. was that this code be by the four of and that it the from the of Dayrat and Clarke, (1) an epithet-based form of species name in which the specific name (ICZN) or epithet (BC, ICBN) is combined with the the and the number of the Schander and Thollesson, 1995; Dayrat et al., 2004; Dayrat, 2005; Dayrat and Gosliner, and (2) a for species names Queiroz, that a single is with to species and of new species names that would also be published and available under rank-based codes after the Cantino and de Queiroz the rules governing species names that to the 1998 Harvard Workshop and the to and A of which about the and of this species The four of on May at the of in to these which led to a that species names be dealt with in a Because the led to the mechanism for with species names, we below. the most to a new species code would have been the of two names for every that the draft species code adopted epithet-based form of species name. the code we been drafting to be the species would have two names: under the and under the PhyloCode. This would the and among other of species names. phylogenetic nomenclature does not to change the names of clades Queiroz and Cantino, and the PhyloCode the use of existing clade names clades have the same name under the PhyloCode and the rank-based The of of rules for naming species under the rank-based and phylogenetic to even differences species names under the two The because created epithet-based species names (e.g., would not necessarily the requirements of the rank-based the and requires that a specific epithet be combined with a genus name for a species name to be the name would not be available under the where name must be published in combination with a generic name or by (Article the PhyloCode to its rules for species names, with a form for these names and rules for their that are incompatible with those of the rank-based codes, under one of the rank-based codes to species under the PhyloCode, even with a This would of species names. Under the PhyloCode, conversion of existing clade names (i.e., to the rules of the requires that the name be published and in an that be implemented with the this to the to existing species names 2003) would be it to be there would be a of during which most species names would not be and under this conversion of species names without and registration be but so would an the clade and species names are and the registration is by to be a of the PhyloCode et al., an epithet-based approach to and number as part of the species which was used in draft species also has because more to than words, the and are likely to be in which could to the same epithet is used in species names. also how to names of Some such taxa may be clades in which case can be any other clade the PhyloCode would have to be than but this is taxa may be distinct and the naming of these is that be treated as ranked taxa species would be with the of mandatory ranks in the PhyloCode. these could be as species to be species Queiroz, a from the that every to only one species. from is not in a to reject this approach, but the practical for who use species names, are a in the rank-based codes, species in draft species code was to a single that would as a reference for application of the name. the rules under the and the in some and under the but not under the and the of and A may be as a under the but not under the which only a may be as a and the does not allow of a are the of does not of a under the a a single set of rules for to disagreement about the status of used as for the same species name under the PhyloCode the rank-based do not in and of to the development of a species but it is a is disagreement among the four of about how an this would have Although of these would the development of a species their led to its and the could be we to that such a code would be for the the names of all species is a which would only be there a major On the phylogenetic nomenclature of the manner in which species names in draft species code was not fundamentally from rank-based i.e., species names as the species a in of a of to drafting a species we ultimately the in of an approach that is intermediate an new set of rules to species names and simply in the PhyloCode. Under the new approach that we governance of and of new species names is to the rank-based codes, but are to the PhyloCode that allow those names to be and used in a that is with the of phylogenetic nomenclature. the approach at the second of the ISPN (Yale University, This was by the and the four of with these ideas as a new article in the PhyloCode. This article (Article 21) was prepared the and to the CPN for In the of the Wolsan, a of the an in Wolsan, from the one in only one a species name adopted in the context of phylogenetic nomenclature would take the form of the (ICZN) or species binomen the and would be permanently In under the approach the species name that is currently under the rank-based code would be Although the combination would species names, it would in a and of the species names under phylogenetic nomenclature. discussion of the CPN in May and adopted the one The new has four (1) requires with the rank-based codes for the and of new species names. (2) The two parts of Linnaean binomina are in a that is with the of phylogenetic nomenclature, most of mandatory binomina are under the rank-based codes, the first of a binomen is as a species than as a genus name. of a species is not the species name is a is it is not to any rank under the PhyloCode. are to the of genus names under the rank-based codes in a that is compatible with clade names governed by the PhyloCode, and are to phylogenetic information about the genus names explain how the incompatibility Linnaean binominal species names and phylogenetic nomenclature. The is available in the most version of the PhyloCode but some are for provides a mechanism for species names under the rank-based codes in the context of phylogenetic nomenclature. which is used in the PhyloCode, is used to and The was to the incompatibility species nomenclature which the use of the genus rank is and phylogenetic nomenclature which ranks are not In order to this a but important how names are governed and how are that the PhyloCode does not or of species names. rules of the rank-based codes (BC, must be a new species name is published and In order that these names be compatible with phylogenetic nomenclature (e.g., that the use of ranks not be must be The of species names by consists of any mandatory reference to genus and species Thus, it this code is of the first part of a species binomen is not as a genus name but as simply the first part of the species name and the second part of a species binomen is associated with the species as a of biological not as a and species are as of biological under the PhyloCode. are not ranks because do not to in a Under the PhyloCode, a clade is as a of an or and all of its a species is as a of a that is from other such of how it is (e.g., by of and species may be for the species is the clade which is the clade PhyloCode in all However, and do not to ranks of a (e.g., genus for for In some the names of a clade and a species may to the same set of under However, those two names would the clade name cannot be a of the species and Because the clade all the but the species does not the names would no to the same set of the species to or the set of to of more than one species. the biological entity to taxa that are associated with ranks and the of an name is associated with the species than the rank of nomenclature. Thus, names under the rank-based may be used to to separated species The of and its are to the PhyloCode to as a and to for a a new species for its nomenclatural status under the PhyloCode, and for some phylogenetic information about is the first part of a species A has no necessary to any rank under this However, to the requirements of the rank-based codes, a must be used or associated with the rank of a new species A used a new species name not have been as a clade name under the PhyloCode. the status of under the a new species name under the rank-based some mechanism should be used to whether the generic name is an clade name under this This status may be in a or by the use of The does not a set of but two are one of which no could that this has been as a clade name under this could that the is not an clade name governed by a rank-based one use to situations (e.g., and the latter governed by the of a symbol to nomenclatural status is because its may simply from of the some may of a symbol because are of the The of are would be in the of the is common to with this used for a the use of a symbol to that the name is not an clade name would be provides for a for use as a generic name a new species When the name of a new of a generic name consideration of the nomenclatural under the rank-based code and this In a generic name (BC, ICBN) or name (ICZN) that is also an clade name (or is being as a clade name) under this code should be the names of more than one clade in a of clades these any one of the names may be this is not an existing generic (or name may be even the of the associated taxon under the rank-based code is or or a new generic name may be the species to be cannot be to any taxon with which a generic (or name has been associated under the rank-based the only is to a new name to as a generic name under the rank-based code under this This name may be as a clade name under this In cases of phylogenetic uncertainty, the are the same as under the rank-based codes a new species (1) a new name or (2) use an existing name for a taxon that is of or The of with one of these two in the of is a by the rank-based codes and a of the approach to species names adopted in the PhyloCode. In to the rank-based codes, the PhyloCode for information about the phylogenetic status of the taxon used as the which we the for naming a new to the reference or of that is a clade name (Article is to the of to that by the rank-based Although the and methods for species have in and new species are described without these whether a is based on a unique combination of is not based on of are not and the to which a set of is as a new species are not that the to a new species by to other to should be a new species name under the rank-based the should a of the that the species represents a from other or an to a this The PhyloCode is with to species and are not to the species and at The that new species names used in the context of the PhyloCode be under the rank-based codes that are also as available (ICZN) or published (BC, ICBN) under those However, a name is available or it may be used in various that are more with the of phylogenetic nomenclature. this some that the use of uninominal species names (Cantino et al., 1999a; Dayrat et al., 2004; Dayrat, 2005). to a species binomen available (ICZN) or published under the rank-based the second part of the species binomen may be treated as the name of the species (i.e., a species under this In this the species uninomen may be combined with the names of clades other than the When the second part of a binomen under a rank-based is used as the name of the it is that it be by the and/or the and of the specific epithet (BC, ICBN) or name the species that is referred to as or under the be referred to under the PhyloCode as or or that the is not used in combination with the specific name or the and should be the is of the and are discussed in the may be used to whether the is an clade name or the taxon to which it is to be but such are from the in this for some in species nomenclature by to specific or names with clade names that are ranked the genus under the codes after of the species names. codes to assign every species to a taxon of genus rank, even in cases in which about phylogenetic In such to a genus may be to to other genera and of phylogenetic by a clade name for the name of a nonmonophyletic or monotypic genus, is more with the of phylogenetic nomenclature than maintaining the that every species to a is sometimes to a uninominal species name with more than one or with names of more clades, as a mechanism for phylogenetic The relationships among these taxa can be in a of but for the of the PhyloCode the names of such taxa in order of from to the species has been in at least two and those names to be under the PhyloCode as the names of clades, the name and relationships of the species could be in any of the is not or or the name of a species under the is and has not been as a clade name under the PhyloCode because there is to and the name has been as the name of a more clade under the PhyloCode, the name and relationships of the species could be in any of the is not or or or the use of names, which the rank-based that a specific name or epithet was combined with a genus name. in the name the that was combined with a genus name other than in the Under the rank-based codes, this use of is mandatory the is In under the PhyloCode, the use of in such situations is even the is Because the PhyloCode is of information about the changing of specific names or with particular of generic rank is not as important as it is in the rank-based The PhyloCode specific names or to be combined with various clade names, including names associated with the genus (e.g., and see in rank-based codes, names associated with ranks other than genus in those codes (e.g., and names not associated with any rank. important for addressing the use of species names in the context of Phylogenetic Nomenclature is that species may be used as in phylogenetic the application of clade names in the of phylogenetic are in of the PhyloCode. When used as the names of species can be in any of the described as in the of this and with (or the discussed to additional information about the that as clade with the first or species to a of the of the as by that is an clade name is not the taxon is to be this the species used as a in this could be in any of the additional (not an or or or or phylogenetic species names are to new genera under rank-based nomenclature. likely the names of species that been used as in phylogenetic this of nomenclatural could be as an in of species in the context of phylogenetic a of and of as as new under rank-based nomenclature, the names of some species used as of other names. The to this is in which a species is as a the is the of that species Thus, currently species the of the species name in the is the the species name as a is no either because the species has been or to a genus, the species name with which it has been the name of the species. However, the of the name used as a than that of the currently species the that the discussed does not are used as (as by However, as may be naming clades of species and The article on species names recently adopted in the PhyloCode a major of that code and has important (1) names used in the context of phylogenetic nomenclature be the rank-based code and be published or available under that (2) The PhyloCode is in addressing clade and species names. the to and as as existing species names (as would be the case the PhyloCode governed species who wish to use the PhyloCode and species names to do so in a that is with the of phylogenetic nomenclature. phylogenetic information by to whether names are clade names and whether taxa to which various names are monophyletic or and by in the specific names or are combined with supraspecific names. nomenclatural and of phylogenetic information by combining the most of most the use of single as reference for the application of names, with critical in the specific names or can be combined with supraspecific names and in by that the used for species. In so this approach represents an in species nomenclature and a major to of the PhyloCode. The development of the ideas in this article has from discussions with are to the participants in the Second of the ISPN (Yale University, 2006), where the of first and and